ICML-98 Submission #67

The Kernel-Adatron: A Fast and Simple Learning Procedure 
for Support Vector Machines

Thilo Friess 
Dept. of Automatic Control and Systems Engineering
University of Sheffield, UK

Nello Cristianini 
Colin Campbell 
Engineering Mathematics Department 
University of Bristol, UK

ABSTRACT: 

Support Vector Machines work by mapping training data for
classification tasks into a high dimensional feature space. In the
feature space they then find a maximal margin hyperplane which
separates the data. This hyperplane is usually found using a quadratic
programming routine which is computationally intensive, and is of non
trivial implementation.

In this paper we propose an adaptation of the Adatron algorithm for
classification with kernels in high dimensional spaces. The algorithm
is simple to implement and finds a solution very rapidly with an
exponentially fast rate of convergence (in the number of iterations)
towards the optimal solution. Experimental results with real and
artificial datasets are provided.


Keywords: Support Vector Machine, Large Margin Classifier, Adatron,
Statistical Mechanics



Contact author data: 

Thilo Friess 
Department of Automatic Control and Systems Engineering
University of Sheffield 
Mappin Street 
Sheffield, S1 3JD,  UK 
Telephone +44 (0)114 222 5250 
Fax +44 (0)114 273 1729